Abstract

The mental health and wellbeing of various professions is an important predictor of economic outcomes, workforce retention as well as a broader indicator of our socioeconomic priorities. Recent changes and challenges to the nature of professional work in Australia, including shifting employee expectations, such as lockdowns and the rise of remote work due to the global COVID-19 pandemic, have highlighted the connection between psychosocial job characteristics and mental health. Here we describe the connection between mental health and job demands, job control and job security, and how it has changed in the past 20 years in Australia using annual nationally representative survey results from HILDA.



Methods


Variable definitions


Professional codes

Professions were defined by the ISC 4-digit codes (i.e., jbm688) available in the restricted HILDA dataset, based on the Australian and New Zealand Standard Classification of Occupations (ANZSCO 2006).


Remoteness

The Accessibility/Remoteness Index of Australia (ARIA+) is provided in HILDA (hhsra) and based on the Australian Statistical Geography Standard Remoteness Area framework (Summerfield et al., 2021). We defined the remoteness of the region of each teacher and collapsed the three most remote categories into a single category to produce three levels of remoteness: Major Cities, Inner Regional and Other (Outer regional, remote and very remote).


Psychosocial characteristics of work

The psychosocial characteristics of work as defined by Butterworth, Strazdins, Rodgers, & Leach (2010), included three psychosocial job components (factors) from 12 items in all waves as well as an additional nine items available from Wave 5 (2005). We used all available items in each wave and constructed scores for each component by calculating the average of item responses (after reverse scoring the negatively worded items). Each component and the corresponding items are shown below. Item scores range from 1 to 7.


Job demands & complexity

code description waves
jomms My job is more stressful than I had ever imagined all waves
jompi I fear that the amount of stress in my job will make me ill all waves
jomcd My job is complex and difficult all waves
jomns My job often required me to learn new skills all waves
jomus I use my skills in current job all waves
jomini My job requires me to take initiative 5:20
jomfast I have to work fast in my job 5:20
jomwi I have to work very intensely in my job 5:20
jomtime I don’t have enough time to do everything in my job 5:20

Job control

code description waves
jomfd I have freedom to decide how I do my own work all waves
jomls I have a lot of say about what happens in my job all waves
jomfw I have freedom to decide when I do my work all waves
jomdw I have a lot of choice in deciding what I do at work 5:20
jomflex My working times can be flexible 5:20
jombrk I can decide when to take a break 5:20
jomrpt My job requires me to do the same things over and over again 5:20
jomvar My job provides me with a variety of interesting things to do 5:20

Job security

code description waves
jompf I get paid fairly for the things I do in my job all waves
jomsf I have a secure future in my job all waves
jomcsb Company I work for will still be in business in 5 years all waves
jomwf I worry about the future of my job all waves



Check sample

# Count by profession
occupations %>%
  count(year, profession) %>%
  spread(profession, n, fill = 0) %>%
  mutate(Total = rowSums(select(., -year)))
## # A tibble: 18 × 5
##     year Teachers Nurses Accountants Total
##    <dbl>    <dbl>  <dbl>       <dbl> <dbl>
##  1  2005      349    180         110   639
##  2  2006      336    203         114   653
##  3  2007      350    199         109   658
##  4  2008      332    192         112   636
##  5  2009      354    194         103   651
##  6  2010      350    186         100   636
##  7  2011      434    244         153   831
##  8  2012      403    235         148   786
##  9  2013      423    231         144   798
## 10  2014      393    259         141   793
## 11  2015      403    238         134   775
## 12  2016      389    252         131   772
## 13  2017      428    246         129   803
## 14  2018      426    255         139   820
## 15  2019      419    259         127   805
## 16  2020      406    274         128   808
## 17  2021      410    272         135   817
## 18  2022      413    270         114   797


# Count by region
occupations %>%
  count(year, hhsra) %>%
  mutate(hhsra = fct_relevel(hhsra, "Major city")) %>%
  spread(hhsra, n, fill = 0) %>%
  mutate(Total = rowSums(select(., -year), na.rm=T))
## # A tibble: 18 × 7
##     year `Major city` `Inner regional` `Outer regional` Remote `Very remote` Total
##    <dbl>        <dbl>            <dbl>            <dbl>  <dbl>         <dbl> <dbl>
##  1  2005          433              144               56      3             3   639
##  2  2006          445              140               61      4             3   653
##  3  2007          433              154               60      9             2   658
##  4  2008          426              141               58      7             4   636
##  5  2009          433              142               62     10             4   651
##  6  2010          440              132               52      8             4   636
##  7  2011          582              160               73     12             4   831
##  8  2012          550              164               60      9             3   786
##  9  2013          564              168               57      8             1   798
## 10  2014          576              152               58      5             2   793
## 11  2015          546              155               62      6             6   775
## 12  2016          548              147               68      6             3   772
## 13  2017          561              152               79      8             3   803
## 14  2018          579              159               66     11             5   820
## 15  2019          561              162               70      8             4   805
## 16  2020          568              155               77      3             5   808
## 17  2021          560              164               81      7             5   817
## 18  2022          551              158               75      8             5   797



Survey results


Table 1. Change in demographics between 2005 and 2020

Group

Characteristic

20051

20221

p-value2

Teachers

Total

N = 349

N = 413

Female

253 (72%)

293 (71%)

0.6

Age

44 (35, 51)

41 (32, 52)

0.2

Coupled

260 (74%)

319 (77%)

0.4

New parent

50 (14%)

79 (19%)

0.078

Edu

<0.001

Postgraduate

27 (7.7%)

87 (21%)

Graduate diploma

119 (34%)

94 (23%)

Bachelors degree

126 (36%)

178 (43%)

Year 12

75 (21%)

54 (13%)

Year 11 or below

2 (0.6%)

0 (0%)

Tenure (years)

9 (3, 18)

8 (3, 17)

0.3

Region

0.10

City

225 (64%)

279 (68%)

Regional

91 (26%)

83 (20%)

Remote

33 (9.5%)

51 (12%)

Real household income ($000s)

61 (47, 73)

76 (60, 97)

<0.001

Mental health

80 (68, 88)

76 (64, 84)

<0.001

Life satisfaction

8.00 (7.00, 9.00)

8.00 (8.00, 9.00)

0.4

Nurses

Total

N = 180

N = 270

Female

165 (92%)

244 (90%)

0.6

Age

42 (35, 48)

38 (29, 53)

0.2

Coupled

132 (73%)

192 (71%)

0.6

New parent

35 (19%)

43 (16%)

0.3

Edu

<0.001

Postgraduate

4 (2.2%)

32 (12%)

Graduate diploma

42 (23%)

63 (23%)

Bachelors degree

84 (47%)

139 (51%)

Year 12

38 (21%)

36 (13%)

Year 11 or below

12 (6.7%)

0 (0%)

Tenure (years)

5 (2, 13)

6 (2, 13)

0.3

Region

0.8

City

115 (64%)

180 (67%)

Regional

42 (23%)

60 (22%)

Remote

23 (13%)

30 (11%)

Real household income ($000s)

58 (47, 72)

79 (61, 96)

<0.001

Mental health

80 (68, 84)

76 (64, 84)

0.012

Life satisfaction

8.00 (7.00, 9.00)

8.00 (7.00, 9.00)

0.11

Accountants

Total

N = 110

N = 114

Female

40 (36%)

61 (54%)

0.010

Age

38 (29, 50)

39 (32, 48)

0.7

Coupled

83 (75%)

93 (82%)

0.3

New parent

15 (14%)

24 (21%)

0.14

Edu

0.003

Postgraduate

5 (4.5%)

19 (17%)

Graduate diploma

20 (18%)

26 (23%)

Bachelors degree

58 (53%)

56 (49%)

Year 12

23 (21%)

13 (11%)

Year 11 or below

4 (3.6%)

0 (0%)

Tenure (years)

4 (1, 10)

4 (2, 10)

0.7

Region

0.7

City

93 (85%)

92 (81%)

Regional

11 (10%)

15 (13%)

Remote

6 (5.5%)

7 (6.1%)

Real household income ($000s)

68 (55, 89)

81 (62, 106)

0.005

Mental health

80 (68, 84)

76 (68, 84)

0.4

Life satisfaction

8.00 (7.00, 8.00)

8.00 (7.00, 9.00)

0.2

1N = N; n (%); Median (IQR)

2Pearson's Chi-squared test; Wilcoxon rank sum test; Fisher's Exact Test for Count Data with simulated p-value
  (based on 2000 replicates)



Mental health

Mental health was measured by the MHi-5 score (0-100) and compared to life-satisfaction responses (multiplied by 10 to match 0-100 range).


Figure 1. Mean ±SE mental health (dark) and life-satisfaction (light) by profession over time



Figure 2. Mean ±SE mental health by region and profession over time



Figure 3. Mean ±SE mental health by school over time



Psychosocial characteristics of work

Job demands, job control and job security components were calculated from the averages of the items in each respective table (shown above).


Figure 4. Mean ±SE psychosocial job components by profession over time



Figure 5. Mean ±SE psychosocial job components by school over time



References

Butterworth, P., Strazdins, L., Rodgers, B., & Leach, L. (2010). Deriving an evidence-based measure of job quality from the HILDA survey. Australian Social Policy, (9), 67–86.
Summerfield, M., Garrad, B., Hahn, M., Jin, Y., Kamath, R., Macalalad, N., et al.others. (2021). HILDA user manual–release 20. Melbourne Institute of Applied Economic and Social Research, University of Melbourne.